A novel, sequence-independent strategy has been developed for the amplification of full-length cDNA copies of the genes of double-stranded RNA (dsRNA) viruses. Using human (Bristol) group C rotavirus as an example, a single amino-linked modified oligonucleotide (primer 1) was ligated to either end of each dsRNA genome segment by using T4 RNA ligase. Following reverse transcription, annealing, and repair of cDNA strands, amplification of the viral dsRNA genome was accomplished by polymerase chain reaction using a single complementary oligonucleotide (primer 2). Northern (RNA) hybridization of cDNA to virus dsRNA indicated that it was possible to generate cDNA representing the complete genome from very small clinical samples. This technique was used to determine the complete nucleotide sequence (728 bp) and coding assignment of gene 10, which revealed an open reading frame of 212 amino acids with limited homology to NS26 from human group A rotavirus. In contrast to previous tailing methods, the addition of one defined primer allowed unequivocal identification of terminal nucleotides and should be generally applicable to viruses with segmented dsRNA genomes and especially for analysis of clinical samples, for which very limited quantities of biological material are available.

~ ~~The sequences of the por genes, encoding outer-membrane protein PI, have been obtained from a number of strains of Neisseria gonorrhoeae that express PIA molecules with Mering serovar specificities. The inferred amino acid sequences of the mature proteins each comprise 308 residues and show considerable homology, with the degree of sequence variation between PIA molecules being considerably less than seen previously with PIB, but more evenly distributed throughout the molecule. The positions of sequence variation are largely confined to the regions predicted to form one of eight surface-exposed loops, suggesting a more widespread distribution of potential antigenic diversity. The deduced amino acid sequences were used to synthesize peptides for epitope mapping experiments. Some epitopes responsible for serovar specificity or recognized by bactericidal monoclonal antibodies could be identified on the basis of their reactivity with simple linear peptides, whilst others recognized conformational epitopes. By comparison of sequence differences with mAb reactivity it was possible to identify regions that appear to contribute to such determinants, including separated regions of the molecule which together were required for the formation of the conformational epitopes. All the epitopes identified lie at or close to the apices of the predicted surface-exposed loops 1,3,6 or 8, focusing attention on these regions as accessible targets for immune attack.

The gene coding for the major inner capsid protein VP6 of human group C rotavirus was cloned into baculovirus using the pBlueBac2 vector and expressed in insect cells. When cultured in High Five cells, VP6 was expressed at a high level and exported to the cell culture medium. Purified VP6 was used to immunise rabbits. Hyperimmune rabbit serum, which reacted with native human group C rotavirus in infected cells, was used to develop and optimise an EIA for the detection of antibodies to group C rotavirus using the recombinant VP6 as a source of antigen. In a local epidemiological survey of 1000 sera grouped by age, an average of 43% of samples were found to have antibodies to human group C rotavirus with the highest proportion (66%) in the 71-75 year age group. In comparison, 97% of adults and 85% of children had antibodies to recombinant VP6 from the bovine RF strain of group A rotavirus. These results suggest that infection with human group C rotavirus is a common occurrence despite the apparent rarity of reports of human group C rotavirus in clinical samples from patients with gastroenteritis.